This project analyzes the role of somatosensory neurons in the parietal lobe during performance of skilled manual tasks. It aims to understand how the hand acquires information about objects through the senses of touch and proprioception, and uses it to grasp and manipulate them. We use a prehension task, in which the hand grasps and manipulates objects, as a model system to examine how sensory cues and previous experience are used to plan and implement skilled hand behaviors. We hypothesize that during active touch, internal representations of the sensory inflow are implemented by corollary discharge from the motor system so that the subject can predict the sensory consequences of intended actions. Convergence of central and peripheral signals allows neurons in posterior parietal cortex (PPC) to compare predictions and reality. The sensory responses are perceived in the context of task goals. Multiple electrode recordings of spike trains and local field potentials in S-I and PPC, and measurements of hand kinematics, assess temporal relations between neural populations representing the fingers. We propose that anticipatory precontact activity in PPC reflects task planning needed to grasp objects efficiently and to secure them for manipulation. Post-contact activity in S-I confirms or rebuts the subject[unreadable]s expectation of haptic features, and provides feedback needed for error correction. Aim 1 analyzes top-down cognitive control of sensory feedback by varying information provided by the cue. We examine how the shape and location of an object are represented in PPC when it is grasped with different instructions and expectations. Aim 2 examines the neural control of bilateral hand movements by comparing temporally uncoupled grasping actions performed by the left and right hands, with similar movements that require coordinated, and synergistic actions of the hands. Using simultaneous bilateral recordings from left and right hemispheres, we examine whether such behaviors involve neurons in each hemisphere specialized for bimanual actions, or are implemented by synchronous activation of the two hemispheres. Aim 3 explores the role of PPC in decision making when cues are ambiguous. We assess the role of handedness preferences, reward probability and short-term memory in choice of the hand used and object grasped in each trial. This research provides basic insights into the dynamic organization of cortical circuits, the role of prediction in normal hand use, and integration of somatosensory information between hemispheres needed for fine motor control of the fingers. An understanding of these cortical processes may have clinical importance for rehabilitation following neurological disorders such as stroke or peripheral nerve injury. Principles of sensorimotor integration derived from this research may prove useful for developing better sensory prostheses or robotic manipulators based on biological models of hand function.